Bay-region diol epoxide (DE) metabolites of carcinogenic polycyclic aromatic hydrocarbons such as benzo[a]pyrene (BaP) are believed to initiate cell transformation by formation of covalent DNA adducts. Two diastereomeric DEs, each of which exists as a pair of enantiomers, are formed metabolically in mammals: DE-1, in which the benzylic hydroxyl group and epoxide oxygen are cis, and DE-2, in which these substituents are trans. BaP DE adducts described in the present report are all derived from the latter diastereomer. The major adducts formed in DNA by these DEs result from cis and trans opening of the epoxide ring by the exocyclic N-2 and N-6 amino groups of deoxyguanosine (dG) and deoxyadenosine (dA), respectively. We have developed powerful new synthetic methods that make it possible to prepare site- and stereospecific DE-DNA adducts for use as tools to probe the structural biology and mechanistic enzymology of enzymes required for essential cellular functions involving DNA processing. These enzymes include a variety of DNA polymerases such as the newly discovered Y superfamily (see below) of lesion-bypass polymerases, pol zeta, the gap-filling polymerase pol beta, and HIV-1 reverse transcriptase, as well as topoisomerases, helicases and HIV-1 integrase. Studies during the past year have focused mainly on the processing of DE-DNA lesions by polymerases belonging to the Y family. These enzymes, in contrast to the highly accurate replicative polymerases, are less severely blocked by bulky DNA lesions and can replicate past these adducts, albeit often in an error-prone manner. Kinetic studies of BaP DE lesion bypass by the SOS-induced Y-family polymerases, pols IV and V of Escherichia coli, indicated that pol V was the only SOS-induced polymerase capable of base insertion and extension beyond BaP DE-dA lesions. It exhibited limited fidelity, with the most common error (~10-20% relative to correct T incorporation) being A misincorporation opposite the adduct. With BaP DE-dG adducts, pol V was much more error-prone and inserted an incorrect A up to 350 times more frequently than the correct C. In contrast, pol IV was the most efficient and accurate in replicating past BaP DE-dG lesions. Pol IV was also implicated in bypass of a BaP DE-dG adduct in vivo: M13 phage containing a dG adduct was replicated ~4-fold less efficiently in a mutant strain of E. coli lacking pol IV than in wild type E. coli. We have also examined base insertion and extension opposite BaP DE- dG adducts by human DNA polymerase eta. This enzyme was highly inaccurate in base insertion opposite four stereoisomeric BaP DE-dG adducts (derived from cis and trans opening of the two BaP DE-2 enantiomers), and showed a strong preference for purine misincorporation. A possible structural basis for this observation involves preferential purine misinsertion opposite dG adducts with the unusual syn glycosidic orientation, consistent with the known syn conformation (from 2-dimensional NMR studies) of such an adduct at a template-primer junction. BaP DE-dG adducts constituted a significant block to pol eta such that further extension beyond the adduct site was poor. In contrast, yeast pol zeta (a B-family DNA polymerase) readily carried out extension beyond each of the two diastereomeric trans opened BaP DE-2-dG adducts (derived from the DE enantiomers), and was also highly accurate, inserting exclusively the correct C opposite these adducts. This suggests a possible role for eucaryotic pol zeta in continuing the replication of DNA whose extension by the Y-family polymerases is blocked immediately after incorporation of a base opposite the adduct.